Výsledky vyhledávání - "Diaz, Víctor"

Report

From Neurons to Neutrons: A Case Study in Interpretability

Autor: Kitouni, Ouail, Nolte, Niklas, Pérez-Díaz, Víctor Samuel, Trifinopoulos, Sokratis, Williams, Mike

Mechanistic Interpretability (MI) promises a path toward fully understanding how neural networks make their predictions. Prior work demonstrates that even when trained to perform simple arithmetic, models can implement a variety of algorithms (someti

Externí odkaz: http://arxiv.org/abs/2405.17425

Zobrazit plný text záznamu

Report

Unsupervised Machine Learning for the Classification of Astrophysical X-ray Sources

Autor: Pérez-Díaz, Víctor Samuel, Martínez-Galarza, Juan Rafael, Caicedo, Alexander, D'Abrusco, Raffaele

The automatic classification of X-ray detections is a necessary step in extracting astrophysical information from compiled catalogs of astrophysical sources. Classification is useful for the study of individual objects, statistics for population stud

Externí odkaz: http://arxiv.org/abs/2401.12203

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání